INFORMATION RETRIEVAL SYSTEM WITH CUSTOMIZATION

• US 20110264665A1
• Filed: 04/26/2010
• Published: 10/27/2011
• Est. Priority Date: 04/26/2010
• Status: Abandoned Application

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Mital, Vijay, Rubin, Darryl Ellis, Bergstraesser, Thomas Frank
• Application Number:
  US12/767,247
• Publication Number:
  US 20110264665A1
• Time in Patent Office: Days
• US Class Current: 707/741
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

A data search and retrieval system that, in response to a search query, dynamically selects and applies a model of information to be returned to a user. The model may be selected based on the search query directly or indirectly based on data returned by a search engine applying the query. For this purpose, the system may include an index of models, similar to a search index. Models may be authored and contributed to the search and retrieval system by third parties, and an association between each such contributed model and characteristics of a search query, such as specific search query terms, may be stored in the index of models. A user of the search and retrieval system may provide feedback on a model that was used to generate information in response to the user'"'"'s search query, and such feedback may be used to update the index of models.

Claims

1. A search and retrieval system comprising:

at least one processor for implementing a plurality of computer-executable components, the components comprising;

an authoring component adapted to generate a model based on input from a user, wherein the model defines computations to be performed on search results; and

an indexing component adapted to store in an index an association between the model and characteristics of a search query.

2. The search and retrieval system of claim 1, further comprising a computer storage medium, and wherein the authoring component is further adapted to store the generated model in the computer storage medium.

3. The search and retrieval system of claim 1, wherein receiving a model by the authoring component comprises receiving over the Internet a file comprising declarative statements.

4. The search and retrieval system of claim 1, wherein generating a model by the authoring component comprises:

receiving user input comprising declarative statements via the authoring component; and

generating the model based on the declarative statements.

5. The search and retrieval system of claim 1, wherein:

the user comprises a first user and the model comprises a first model; and

the authoring component is further adapted to receive input from a second user and to generate a second model based on the input from the second user.

6. The search and retrieval system of claim 5, wherein:

the indexing component is further adapted to store an association between the second model and the characteristics of a search query.

7. The search and retrieval system of claim 5, wherein:

the characteristics of a search query comprise characteristics of a first search query;

the association comprises a first association; and

the indexing component is further adapted to store in the index a second association between the second model and characteristics of a second search query.

8. The search and retrieval system of claim 7, further comprising:

a model selection component adapted to select one of the first model and the second model based at least in part on a comparison between a query received as input from a third user and the first and second associations stored in the index;

a model application engine adapted to apply the selected model on search results returned in response to the query received as input to generate information.

9. At least one non-transitory computer storage medium comprising computer-executable instructions that, when executed by at least one processor, perform a method within a search stack in a search system, the method comprising:

receiving feedback from a user related to a declarative model that has been used to generate information returned to the user in response to a search query; and

based on the received feedback, modifying an association stored in an index between the declarative model and characteristics of the search query.

10. The at least one non-transitory computer storage medium of claim 9, wherein receiving feedback from a user comprises receiving an explicit indication from the user of the relevance of the declarative model to the search query.

11. The at least one non-transitory computer storage medium of claim 9, wherein:

the declarative model comprises a first declarative model; and

receiving feedback from a user comprises receiving user input requesting that a second declarative model be used to generate information instead of the first declarative model.

12. The at least one non-transitory computer storage medium of claim 9, wherein receiving feedback from a user comprises monitoring at least one user behavior indicator selected from the group consisting of a session length in which the user navigates the generated information, a number of clicks the user makes in the generated information, and page exit paths.

13. The at least one non-transitory computer storage medium of claim 9, wherein:

the user comprises a first user;

the method further comprises receiving feedback from a second user related to the declarative model; and

modifying the association stored in the index between the declarative model and characteristics of the search query is performed based on an aggregation of the feedback received from the first user and the feedback received from the second user.

14. The at least one non-transitory computer storage medium of claim 9, wherein:

the declarative model comprises a first declarative model; and

modifying the association stored in the index between the declarative model and characteristics of the search query comprises associating terms of the search query with a second declarative model instead of with the first declarative model.

15. The at least one non-transitory computer storage medium of claim 9, wherein:

the index stores association scores, each respective association score indicating an association between a declarative model and a term;

the characteristics of the search query comprise a plurality of terms of the search query; and

modifying the association stored in the index between the declarative model and characteristics of the search query comprises, for each of the plurality of terms of the search query, modifying a respective association score indicating an association between the term and the declarative model.

16. A method of performing a search using a search system, the method comprising:

operating at least one processor to perform;

providing a declarative model to the search system;

sending a search query to the search system;

receiving information from the search system returned in response to the query, at least a portion of the information being generated by the search system by applying the declarative model; and

providing to the search system feedback related to the usefulness of the applied declarative model to the search query.

17. The method of claim 16, wherein providing a declarative model to the search system comprises:

using an authoring tool executing on the at least one processor;

receiving user input comprising declarative statements; and

generating a file to represent the model; and

sending the file to the search system over the Internet.

18. The method of claim 16, wherein providing a declarative model to the search system comprises sending user input comprising declarative statements to the search system.

19. The method of claim 16, wherein providing to the search system feedback related to the usefulness of the applied declarative model to the search query comprises sending to the search system a choice selected by a user from a plurality of choices presented to the user in order of relevance to the search query.

20. The method of claim 16, wherein providing to the search system feedback related to the usefulness of the applied declarative model to the search query comprises sending to the search system feedback that was explicitly provided by a user.